Saturday, December 26, 2015

The problem with momentum trading - it's good until it's not

There has been a growing research focus on the problem with momentum trading. Momentum trading is subject to crashes, or to put is simply, it's good until it's not. The data is suggestive of two major problems with momentum. First, you may actually only receive higher returns because you are taking on the potential for crash risk. Second, there may be periods when momentum just does not work.

The charts below are from Patrick O'Shaughnessy's research on momentum. First, there are periods of poor performance. You cannot get away from it. It looks like these crashes come after a big market downturn, but there is limited information to classify these events.

Second, if you look at high versus low momentum you will find that are extended periods when momentum will underperform. Currently, we are in one of those periods. If you avoided momentum since the Financial Crisis, you would have been better off.

Is momentum one more strategy that has served its time and should now be put to rest? Yes, if you use a simple momentum approach. The studies of momentum crash risk do not include a risk management component. This is why risk management is so important with trend-following and momentum models.

The main job of the risk management rules is to mitigate the crash risk inherent to these strategies. No investor or manager should think about momentum strategies without a risk management exit strategy. Risk management has been more important over the last five year not because of some new view of finance, but because it is critical to save managers when the strategy is not working. Momentum has been good but currently it is important to look beyond this approach.

No comments: